If subjects give stronger reactions to the control questions as compared to the relevant questions, then their answers are considered truthful. In contrast, stronger reactions to the relevant questions indicate to be deception. The absence of a clear difference in physiological responses between the control and relevant questions in either direction, then the results will be inconclusive. This means that the examiner cannot give a definite opinion with respect to diagnosis of truthfulness or deception, which may in the field result in a further test being constructed and administered.
The rationale is that an innocent person will respond as much to the control questions as to the relevant questions. In contrast, the guilt person will show more physiological responses to the relevant questions than to the control questions.
The polygraph session proper consists of four phases. Firstly, in the pre-test-interview phase, the examiner seeks to convince the examinee that the polygraph is infallible and discuss the formulations of both the relevant and control questions in the test. Then, in the test phase which is begun by connecting the physiological recording instrument to the examinee. It comprises a set of questions, which consists of three relevant questions, three control questions and three neutral questions. Before the test phase, it is often preceded by the card test, where the examiner uses the instrument to detect which of a number of cards the examinee is thinking of. The aim is to elicit different responses in truthful person and guilt people. It should increase fear of detection in the guilt suspect and increase confidence in the truthful persons that the polygraph test will reveal their innocence.
After that, the interpretation of the polygraph result will employ the global approach or the numerical scoring approach. In the global approach, the information are combined in some unspecified manner with evaluations of the case facts and the examinee’s behaviour during the test in order to reach an overall decision about the truthfulness of the examinees. On the other hand, the numerical scoring approach is assigned for each of the physiological parameters for question-pair; the score can range from –3 to +3. If the observed reaction is stronger in response to the relevant question, a negative score is given; positive scores are assigned when the reaction is stronger to the control questions and the value 0 is given when no difference is observed.
If the examiner decides the examinee has been deceptive, the examiner tries upon returning to induce a confession of guilt during the post-test-interview phase. This phase is terminated either by a confession or by the examiner’s decision that the examinee will not be able to get a confession.
After consideration of these four phases in the CQT, some critics argued that there were some limitations in the CQT. Firstly, the test is lack of standardized in the procedures. In formulating the questions, Ray Bull (1988) suggested that it was extremely difficult to devise control questions that ensure the eliciting of stronger reactions in an innocent person than would the relevant questions relating to the crime of which they had been accused. This difficulty in selecting adequate control questions may be a reason for the CQT leading to more false positives (classifying truthful people as liars) than false negatives (classifying liars as truthful) (Carroll, 1988).
Another consideration is that Ben Shakhar and Furedy (1990) have pointed out that it is not clear, why a guilt suspect should be less concerned with the control questions. Moreover, Ekman (1992) maintained that there might be cases when the truthful people exhibited stronger responses to relevant questions than the control questions. For example, the examinees may view the police and the polygraph as fallible.
Furthermore, the fact that the CQT is based in part on misleading the suspects implies that its use is rather limited. Many suspects could very easily know the true nature of the control questions. Thus jeopardizing the very logic of the CQT’s inference rule because innocent people familiar with the CQT would not be concerned with the control questions, but with the relevant ones.
Secondly, the scoring of the polygraph charts is subjective which may reduce accuracy.
Bull (1988) said that some polygraph examiners simply looked at the charts and based their conclusions on global, or “eyeball” impressions. Even those who were more precise may still be quite subjective. Many polygraph examiners decided which questions had occasioned the largest responses by merely looking at the charts without bothering to measure each response. They might even use their expectations based on the pre-examination interview, along with the examinee’s physiological reactivity, as determinants of their global classification.
This type of subjectivity is the very contradiction to the scientific measurement model by which psychology seeks objective, replicable observations. Another consideration is that, the training and qualifications of the examiners are also the main problems in the CQT. The examiners might form an impression whether the examinee is guilt before the examination starts, which, in turn, may influence the outcome of the test.
Accordingly, CQT is highly dependent on the examiner concerning not only how the examiner interprets the results, but also concerning what the control questions will be and how the questions are present to the examinees. Thus, a single suspect can undergo very different tests when interrogated by several independent polygraph examiners on the same event. Finally, the CQT is not a standardized and objective test in the psychometric sense, because it depends on a great extent on the examiner and on his or her interaction with the examinee.
After consideration of the limitations in the CQT, it is necessary to evaluate the validity and reliability of the CQT, scientists commonly have two major approaches they can take which are in laboratory experiment and in the field.
Typically, in the laboratory polygraph experiment, subjects are randomly assigned to one of two conditions. In the guilty condition, subjects are committed a mock crime, such as the theft of an object. On the other hand, in the innocent condition, subjects are given a description of the mock crime. All subjects are told to deny the theft, and then given a polygraph test by an examiner.
Raskin, Honts and Kircher have recently reviewed the scientific literature addressing psychophysiological credibility assessment. They found eight laboratory studies in the CQT that consisted of Ginton et al. (1982), Honts et al. (1994), Horowitz et al. (1995), Kircher & Raskin (1998), Podlesny & Raskin (1978), Podlesny & Truslow (1993), Raskin & Hare (1978) and Rovner et al. (1979).
These studies indicated that the CQT was very accurate discriminator of truthful people and deceptive people. The CQT correctly classified about 90 percent of the subjects and produced approximately equal numbers of false positive and false negative error.
On the other hand, the alternative approach to studying psychophysiological credibility assessment is to conduct field studies. In the review from Raskin and his colleagues, they examined the available field studies of the CQT and found four studies of it. These studies consisted of Honts (1994), Honts & Raskin (1988), Patrick & Iacono (1991) and Raskin et al. (1988).
The results of the independent evaluations of the field studies produced results that were quite similar to the laboratory studies. The average accuracy of field decisions for the CQT was 90.5 percent. However, with the field studies nearly all of the errors made by the CQT were false positive errors.
Both laboratory and field studies converge on the conclusion that the CQT is a highly accurate discrimination in truthful and deceptive people. The research converges on an accuracy estimate that exceeds 90 percent when inconclusive outcomes are excluded. Moreover, there may be a tendency for the CQT to produce more false positive than false negative errors.
The laboratory experiment and field studies have been used to evaluate the validity and reliability of the CQT in polygraph. However, there were different supporters and opponents criticized the laboratory and field studies.
Laboratory research has traditionally been an attractive alternative because the scientists can control the environment. Moreover, concerning credibility studies, the scientists can know with certainty who is telling the truth and who is telling a lie (i.e. ground truth) by randomly assigning subjects to conditions. However, these types of studies were lack of realism that may limit the ability of the scientists to apply the results in real-world settings.
Furthermore, the problem with laboratory is the difficulty of inducing in the subjects the degree and type of emotional concern experienced by guilty or by innocent suspects being tested in real life. Lykken also concluded that the laboratory studies that ask volunteer college students to “commit a crime” and lie during an interrogation are creating in more a state of excitement than guilty.
Although field studies are plagued by numerous problems, the chief problem lies in unambiguously determining ground truth. That is, some method is needed for determining who is in fact telling the truth that is independent of the outcome of the test.
High accuracy rate of field studies that was only based on examinees’ responses in polygraph test (blind scoring). This is desirable practice from a scientific point of view, because it eliminates the possible contamination (e.g. counter-measure of the subjects during the examination) in the decisions of the original examiners. However, it is insufficient information of the decision for the legal proceedings.
Another consideration is that the field studies shown a high rate of consistencies between the polygraph expertise. However, in the Floyd Fay case at 1978, these consistencies did not always hold up in real world. Floyd Fay was arrested in 1978 and charged with aggravated murder. He knew he was innocent and decided to take the polygraph test. Several polygraph experts that included of David Lykken, David Raskin, Gordon Barland and Frank Horvath analyzed his polygraph test. Lykken concluded that it was useless and concluded that he was a guilt suspect. Raskin concluded that he was telling the truth and Barland and Horvath concluded that the record was inconclusive. These inconsistencies reflected the serious problems in drawing conclusions from testing procedures that may be unreliable.
In the polygraph test, counter-measures are anything that a subject might do in order to distort or defeat a psychophysiological credibility assessment test. Although deceptive subjects undoubtedly most commonly use counter-measures, innocent subjects may also sometime employ such tactics because of the fear of failing the control question test. Therefore, it cannot automatically be assumed that all subjects who are observed to use counter-measures are ‘guilty’ of the crime they are being questioned about.
There are various types of counter-measure techniques such as by physical, mental, hypnosis, biofeedback and drugs. These are consists of three different ways of defeating the polygraph test. Firstly is suppressing physiological responses to relevant questions. The examinees can make the difference in responsively between the relevant and control questions minimal or in the desired direction.
Secondly is increasing the physiological responses to control questions that can reduce the discriminative power of the relevant questions. Another consideration is that, the examinee can suppress the overall level of physiological reactivity, which make more difficult for the examiner to differentiate between physiological responses to the control and relevant questions.
More recently, laboratory studies reviewed by the Office of Technology Assessment (1986) which have shown that physical counter-measures can be more effectively than other types of methods (such as the use of drugs and mental counter-measure) that utilized by some people in order to defeat the CQT.
These physical counter-measures are biting one’s tongue in response to the control questions may create sufficient pain or discomfort to elicit an artificial physiological response indistinguishable from that of a genuine one. Besides, pressing the toes against the floor or the thighs against the chair the individual is sitting in have been shown to be effective techniques in order to defeat the CQT.
On the other hand, the counter-counter-measures are special techniques employed by polygraph examiners in order to detect subjects who are trying to make deliberate attempts to defeat the polygraph test. The most commonly used in defeat the polygraph test is the physical counter-measure technique.
From the studied of Honts, Raskin and Kircher (1983), they suggested that the plectromyographic recordings obtained from the temporalis and gastrocnemius muscle areas detected 80 percent of subjects using physical counter-measures such as biting one’s tongue. However, from the point of real-life carried out in the field, few polygraph examiners are likely in the near future to have the expertise or equipment to use electromyographic recordings, regardless of the effectiveness in detecting the physical counter-measures. Most examiners will simply rely on close observation of behavioral signs inductive or by careful inspection of the polygraph record itself.
After described the counter-measures and counter-counter-measure in the polygraph test. The application of the polygraph tests in psychophysiological credibility assessment also has seriously impact on the psychological approach and the current legal status. For example, the British Psychological Society (1986) authorized a study of available research literature. It concluded that the evidence supporting the use of polygraph tests was “very slender”, its reliability and validity were in question, and a need existed for more research on the topic, since much of the existing research was inadequate.
Another consideration is that, the United States Supreme Court recently considered the admissibility of polygraph findings in the case of United States v. Scheffer (1988). The appeal challenged the constitutionally of President Bush’s application of Military Rule of Evidence 707, which made the results of polygraph tests inadmissible in all military courts martial. The Justice Clarence Thomas concluded that military rules of evidence call for only reliable evidence to be admitted and that scientists and legal experts are in dispute about the reliability of the polygraph results. His opinion cited a survey of experts by Iacono and Lykken (1997) that concluded accuracy rate to be little above chance. In addition, Honts (1994b) pointed out that polygraph tests have been held inadmissible in courts because it is felt that juries may be overwhelmed by the scientific nature of the evidence they produce (Myers and Arbuthnot). The mock jurors found that the polygraph evidence is the least convincing.
There have been two major technological advances reported in the scientific literature during the last decade. The first of those concerns the forensic use of a technique known as the Directed Lie Test (DLT). The rationale of the DLT predicts that guilt people will respond with greater physiological responses to the relevant questions while the truthful people will response greater in the directed lie questions (i.e. these questions are known lies). The DLT is more standardized and lower rate of false positive errors when compared with the CQT. It is widely used in national security screening settings, and in forensic application.
Another very common finding in the scientific study of decision-making is that statistical decision making usually outperforms expert human decision-makers. So, David Raskin eventually derived a computer-based statistical decision making system for psychophysiological credibility assessment tests in the late 1970s. Given that computer-based statistical analysis is objective and completely reliable. Its widespread use in the field should serve to improve the quality of practice generally and the overall accuracy of results.
Conclusion
The use of physiological recordings to make inferences about the truthfulness of a person’s statement has a long and controversial history in both psychological science and the law. The polygraph measures several physiological reactions. The assumption of the polygraph is that lying is accompanied by psychophysiological arousal that can be objectively monitored by the polygraph.
The most commonly test used for psychophysiological credibility assessment is the control question test (CQT), which involves setting different types of questions and asking the suspects about the crime. However, some criticized the CQT is subjective and not a standardized procedure. In addition, it depends largely on the relationship between the examiner and the examinee. When applied the polygraph test in legal status, if the standards of the polygraph profession can be raised to match the level of scientific achievement, it would seem that polygraph tests might have reached a level of advancement and scientific acceptance, where they might serve as a useful tool in legal proceedings. Although some countries accepted the polygraph test in lie detection, the accuracy rates are too low to justify the outcomes of these analyses being used as the main evidence in court.